A "false alarm" (unwanted alarm) in a fire alarm system is operationally expensive — not only because of the direct interruption, but primarily because of what it does to confidence in the system. When the third false alarm rings this month, nobody runs at the fourth. That is precisely the problem.
In this article: where false alarms typically originate, what they cost, and which technical and organisational measures reduce them.
1. What "false alarm" really means
NEN distinguishes several kinds of unwanted alarm:
- Fault alarm (technical) — detector signals due to fault, defect, contamination
- Unwanted alarm — detector reacts to something that is not a fire (steam, dust, cigarette smoke in a toilet, welding fumes, cooking)
- Inadvertent alarm — manual call point accidentally activated (passer-by, cleaning crew, child)
- Test alarm — deliberately activated for test / drill
The first three we want to minimise; the last belongs at regular intervals.
2. The real cost
A false alarm costs more than you think:
- Time of own staff — typically 1-3 people evaluating the alarm, resetting it, filling out the logbook
- Productivity loss — if evacuation occurred, a building is easily 30-90 minutes unproductive
- Fire service callout if signal transmission to the regional alarm centre is enabled — most municipalities charge for two or more false alarms in a short period
- Cry-wolf effect — at a real fire staff respond more slowly, with greater fire damage as a consequence
A building with >6 false alarms per year has a structural problem that must be resolved — not chronically managed.
3. Typical causes
In projects we analyse these are the top-5 causes:
Wrong detector type for the room. Ionisation detection in a kitchen, optical detection in a room with steam or dust — a recipe for false alarms. The PoR should have intervened here, but often did not.
Contamination. Optical detectors accumulate dust in the sensor chamber. After 5-7 years a contaminated detector gives false alarms on trivial dust or steam triggers. Cleaning or replacement is the fix.
Wrong placement. Detector directly above a desk where welders work, or next to a kitchen exhaust. Physical reality sometimes differs from the engineering drawing.
Inadvertent activation of call points. Often in corridors at places where people bump into them. A call point with a "two-stage action" (open flap + press) reduces this substantially.
Bad renovation handover. A renovation without PoR update: new room layout, old detection strategy. Detector now sits above a glass panel, or right next to a new HVAC outlet.
4. Technical measures
- Multi-sensor detection — detectors that measure smoke + heat + CO and only alarm at combination. Dramatic reduction of false alarms
- Time delay on unstaffed zones — on detection the system waits 60-180 seconds, issues pre-alarm, and only escalates to main alarm on continued detection or a second detector
- ASD with staging — aspirating detection typically has multiple alarm levels (Aware, Action, Fire1, Fire2). The response chain scales with it
- Periodic cleaning — part of NEN 2654 maintenance but often forgotten. Contamination measurement is digitally readable on modern detectors
5. Organisational measures
Technology alone does not solve the problem. A good fire-safety policy also includes:
- BHV confirms alarms before evacuation is initiated — provided response time allows (small buildings, non-care)
- Standard procedures for "risky" activities — hot-work permit required for welding / burning, with prior deactivation of detection in the work area
- Cleaning protocol — cleaning crews use no dust-generating methods under active detectors
- Fire drills not counted in — couple drills to a specific detector zone deactivated in advance, not to a random alarm
6. What NEN 2535 expects in terms of reduction
NEN 2535 does not contain a fixed "maximum false-alarm rate", but inspectors and insurers do apply implicit norms. In practice: more than 3-4 unwanted alarms per year in a 1,000-5,000 m² building is reason for analysis. A thorough inspection report will criticise this.
Closing
A false alarm is not an annoyance but a diagnosis — the system is saying that somewhere there is a mismatch between space use and detection strategy. An installation that does not have unwanted alarms has a good PoR, good engineering and good maintenance behind it.
Do you have a building with recurring false alarms? Schedule an audit — we analyse the last 6-12 months from the logbook and give a concrete reduction plan.